Defeating the Training-Inference Mismatch via FP16

N/A

Defeating the Training-Inference Mismatch via FP16 Penghui Qi*†1,2 , Zichen Liu*1,2 , Xiangxin Zhou*1 , Tianyu Pang1 , Chao Du1 , Wee Sun Lee2 , Min Lin1 1 Sea AI Lab 2 National…

saved by

• Nathan Chen

related reading

• The 4-bitter Lesson | humans&humansand.ai
• [2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMsarxiv.org
• Modifying LLM Beliefs with Synthetic Document Finetuningalignment.anthropic.com
• Is Frontier Asynchronous RL Solved? — Luke J. Huangluk-huang.github.io
• [2502.17424] Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMsarxiv.org
• State of RL for reasoning LLMs | A. Weersaweers.de
• [2604.22082] Removing Sandbagging in LLMs by Training with Weak Supervisionarxiv.org
• Self-Fulfilling Misalignment Data Might Be Poisoning Our AI Modelsturntrout.com
• [2602.17616] Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMsarxiv.org
• [2510.11696] QeRL: Beyond Efficiency -- Quantization-enhanced Reinforcement Learning for LLMsarxiv.org
• Keep the Tokens Flowing: Lessons from 16 Open-Source RL Librarieshuggingface.co
• 2305.14314arxiv.org